[Experimental] Add Kleidi i8mm gemm kernels #1295
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/ao/1295
Note: Links to docs will display an error until the docs builds have been completed. ❗ 1 Active SEVsThere are 1 currently active SEVs. If your PR is affected, please view them below: ✅ No FailuresAs of commit c0ce311 with merge base 26648c2 ( This comment was automatically generated by Dr. CI and updates every 15 minutes. |
facebook-github-bot
added
the
CLA Signed
|
@digantdesai has imported this pull request. If you are a Meta employee, you can view this diff on Phabricator. |
kimishpatel
reviewed
Nov 15, 2024
metascroy
reviewed
Nov 15, 2024
kimishpatel
reviewed
Nov 15, 2024
| @@ -0,0 +1,120 @@ | |||
| // Copyright (c) Meta Platforms, Inc. and affiliates. | |||
There was a problem hiding this comment.
is this entire file codegenable? not for now but just asking
There was a problem hiding this comment.
Yeah good idea, not that much work TBH.
kimishpatel
reviewed
Nov 15, 2024
...al/kernels/cpu/aarch64/kleidi/kai_matmul_clamp_f32_qai8dxp4x8_qsi4c32p8x8_4x8x32_neon_i8mm.h
Show resolved
Hide resolved
kimishpatel
reviewed
Nov 15, 2024
...al/kernels/cpu/aarch64/kleidi/kai_matmul_clamp_f32_qai8dxp4x8_qsi4c32p8x8_4x8x32_neon_i8mm.h
Show resolved
Hide resolved
kimishpatel
reviewed
Nov 15, 2024
...al/kernels/cpu/aarch64/kleidi/kai_matmul_clamp_f32_qai8dxp4x8_qsi4c32p8x8_4x8x32_neon_i8mm.h
Outdated
Show resolved
Hide resolved
kimishpatel
reviewed
Nov 15, 2024
kimishpatel
reviewed
Nov 15, 2024
kimishpatel
reviewed
Nov 15, 2024
metascroy
approved these changes
Nov 15, 2024
kimishpatel
reviewed
Nov 15, 2024
kimishpatel
reviewed
Nov 15, 2024
kimishpatel
approved these changes
Nov 15, 2024
Add kernel level tests, with basic cross compilation support. Tested with S24 + r26c ``` [----------] 6 tests from test_kai_matmul_clamp_f32_qai8dxp4x8_qsi4c32p4x8_8x4x32_neon_i8mm [ RUN ] test_kai_matmul_clamp_f32_qai8dxp4x8_qsi4c32p4x8_8x4x32_neon_i8mm.k_eq_gs_32 [ OK ] test_kai_matmul_clamp_f32_qai8dxp4x8_qsi4c32p4x8_8x4x32_neon_i8mm.k_eq_gs_32 (0 ms) [ RUN ] test_kai_matmul_clamp_f32_qai8dxp4x8_qsi4c32p4x8_8x4x32_neon_i8mm.large_k_n_gs32 [ OK ] test_kai_matmul_clamp_f32_qai8dxp4x8_qsi4c32p4x8_8x4x32_neon_i8mm.large_k_n_gs32 (79 ms) [ RUN ] test_kai_matmul_clamp_f32_qai8dxp4x8_qsi4c32p4x8_8x4x32_neon_i8mm.even_n_gs32 [ OK ] test_kai_matmul_clamp_f32_qai8dxp4x8_qsi4c32p4x8_8x4x32_neon_i8mm.even_n_gs32 (28 ms) [ RUN ] test_kai_matmul_clamp_f32_qai8dxp4x8_qsi4c32p4x8_8x4x32_neon_i8mm.k_eq_gs128 [ OK ] test_kai_matmul_clamp_f32_qai8dxp4x8_qsi4c32p4x8_8x4x32_neon_i8mm.k_eq_gs128 (3 ms) [ RUN ] test_kai_matmul_clamp_f32_qai8dxp4x8_qsi4c32p4x8_8x4x32_neon_i8mm.clamp_k_eq_gs128 [ OK ] test_kai_matmul_clamp_f32_qai8dxp4x8_qsi4c32p4x8_8x4x32_neon_i8mm.clamp_k_eq_gs128 (3 ms) [ RUN ] test_kai_matmul_clamp_f32_qai8dxp4x8_qsi4c32p4x8_8x4x32_neon_i8mm.m_clamp_k_eq_gs128 [ OK ] test_kai_matmul_clamp_f32_qai8dxp4x8_qsi4c32p4x8_8x4x32_neon_i8mm.m_clamp_k_eq_gs128 (5 ms) [----------] 6 tests from test_kai_matmul_clamp_f32_qai8dxp4x8_qsi4c32p4x8_8x4x32_neon_i8mm (121 ms total) [----------] 6 tests from test_kai_matmul_clamp_f32_qai8dxp4x8_qsi4c32p8x8_4x8x32_neon_i8mm [ RUN ] test_kai_matmul_clamp_f32_qai8dxp4x8_qsi4c32p8x8_4x8x32_neon_i8mm.k_eq_gs_32 [ OK ] test_kai_matmul_clamp_f32_qai8dxp4x8_qsi4c32p8x8_4x8x32_neon_i8mm.k_eq_gs_32 (0 ms) [ RUN ] test_kai_matmul_clamp_f32_qai8dxp4x8_qsi4c32p8x8_4x8x32_neon_i8mm.large_k_n_gs32 [ OK ] test_kai_matmul_clamp_f32_qai8dxp4x8_qsi4c32p8x8_4x8x32_neon_i8mm.large_k_n_gs32 (79 ms) [ RUN ] test_kai_matmul_clamp_f32_qai8dxp4x8_qsi4c32p8x8_4x8x32_neon_i8mm.even_n_gs32 [ OK ] test_kai_matmul_clamp_f32_qai8dxp4x8_qsi4c32p8x8_4x8x32_neon_i8mm.even_n_gs32 (28 ms) [ RUN ] test_kai_matmul_clamp_f32_qai8dxp4x8_qsi4c32p8x8_4x8x32_neon_i8mm.k_eq_gs128 [ OK ] test_kai_matmul_clamp_f32_qai8dxp4x8_qsi4c32p8x8_4x8x32_neon_i8mm.k_eq_gs128 (3 ms) [ RUN ] test_kai_matmul_clamp_f32_qai8dxp4x8_qsi4c32p8x8_4x8x32_neon_i8mm.clamp_k_eq_gs128 [ OK ] test_kai_matmul_clamp_f32_qai8dxp4x8_qsi4c32p8x8_4x8x32_neon_i8mm.clamp_k_eq_gs128 (3 ms) [ RUN ] test_kai_matmul_clamp_f32_qai8dxp4x8_qsi4c32p8x8_4x8x32_neon_i8mm.m_clamp_k_eq_gs128 [ OK ] test_kai_matmul_clamp_f32_qai8dxp4x8_qsi4c32p8x8_4x8x32_neon_i8mm.m_clamp_k_eq_gs128 (5 ms) [----------] 6 tests from test_kai_matmul_clamp_f32_qai8dxp4x8_qsi4c32p8x8_4x8x32_neon_i8mm (121 ms total) ```
digantdesai
force-pushed
the
kleidi_i8mm_kernels
branch
from
5f19b69 to
8504177
Compare
digantdesai
force-pushed
the
kleidi_i8mm_kernels
branch
from
8504177 to
c0ce311
Compare
|
Removed operator level test, landing this PR without those. I will debug and add them in another PR as originally planned. Thanks. |
|
@digantdesai has imported this pull request. If you are a Meta employee, you can view this diff on Phabricator. |
digantdesai
added
cpu
topic: not user facing
sunjiweiswift
pushed a commit
to sunjiweiswift/ao
that referenced
this pull request
Nov 25, 2024
* Update git ignore * [experimental] Add Kleidi compile def at the top level * [Experimental] Add Kleidi i8mm gemm kernels Add kernel level tests, with basic cross compilation support. Tested with S24 + r26c ``` [----------] 6 tests from test_kai_matmul_clamp_f32_qai8dxp4x8_qsi4c32p4x8_8x4x32_neon_i8mm [ RUN ] test_kai_matmul_clamp_f32_qai8dxp4x8_qsi4c32p4x8_8x4x32_neon_i8mm.k_eq_gs_32 [ OK ] test_kai_matmul_clamp_f32_qai8dxp4x8_qsi4c32p4x8_8x4x32_neon_i8mm.k_eq_gs_32 (0 ms) [ RUN ] test_kai_matmul_clamp_f32_qai8dxp4x8_qsi4c32p4x8_8x4x32_neon_i8mm.large_k_n_gs32 [ OK ] test_kai_matmul_clamp_f32_qai8dxp4x8_qsi4c32p4x8_8x4x32_neon_i8mm.large_k_n_gs32 (79 ms) [ RUN ] test_kai_matmul_clamp_f32_qai8dxp4x8_qsi4c32p4x8_8x4x32_neon_i8mm.even_n_gs32 [ OK ] test_kai_matmul_clamp_f32_qai8dxp4x8_qsi4c32p4x8_8x4x32_neon_i8mm.even_n_gs32 (28 ms) [ RUN ] test_kai_matmul_clamp_f32_qai8dxp4x8_qsi4c32p4x8_8x4x32_neon_i8mm.k_eq_gs128 [ OK ] test_kai_matmul_clamp_f32_qai8dxp4x8_qsi4c32p4x8_8x4x32_neon_i8mm.k_eq_gs128 (3 ms) [ RUN ] test_kai_matmul_clamp_f32_qai8dxp4x8_qsi4c32p4x8_8x4x32_neon_i8mm.clamp_k_eq_gs128 [ OK ] test_kai_matmul_clamp_f32_qai8dxp4x8_qsi4c32p4x8_8x4x32_neon_i8mm.clamp_k_eq_gs128 (3 ms) [ RUN ] test_kai_matmul_clamp_f32_qai8dxp4x8_qsi4c32p4x8_8x4x32_neon_i8mm.m_clamp_k_eq_gs128 [ OK ] test_kai_matmul_clamp_f32_qai8dxp4x8_qsi4c32p4x8_8x4x32_neon_i8mm.m_clamp_k_eq_gs128 (5 ms) [----------] 6 tests from test_kai_matmul_clamp_f32_qai8dxp4x8_qsi4c32p4x8_8x4x32_neon_i8mm (121 ms total) [----------] 6 tests from test_kai_matmul_clamp_f32_qai8dxp4x8_qsi4c32p8x8_4x8x32_neon_i8mm [ RUN ] test_kai_matmul_clamp_f32_qai8dxp4x8_qsi4c32p8x8_4x8x32_neon_i8mm.k_eq_gs_32 [ OK ] test_kai_matmul_clamp_f32_qai8dxp4x8_qsi4c32p8x8_4x8x32_neon_i8mm.k_eq_gs_32 (0 ms) [ RUN ] test_kai_matmul_clamp_f32_qai8dxp4x8_qsi4c32p8x8_4x8x32_neon_i8mm.large_k_n_gs32 [ OK ] test_kai_matmul_clamp_f32_qai8dxp4x8_qsi4c32p8x8_4x8x32_neon_i8mm.large_k_n_gs32 (79 ms) [ RUN ] test_kai_matmul_clamp_f32_qai8dxp4x8_qsi4c32p8x8_4x8x32_neon_i8mm.even_n_gs32 [ OK ] test_kai_matmul_clamp_f32_qai8dxp4x8_qsi4c32p8x8_4x8x32_neon_i8mm.even_n_gs32 (28 ms) [ RUN ] test_kai_matmul_clamp_f32_qai8dxp4x8_qsi4c32p8x8_4x8x32_neon_i8mm.k_eq_gs128 [ OK ] test_kai_matmul_clamp_f32_qai8dxp4x8_qsi4c32p8x8_4x8x32_neon_i8mm.k_eq_gs128 (3 ms) [ RUN ] test_kai_matmul_clamp_f32_qai8dxp4x8_qsi4c32p8x8_4x8x32_neon_i8mm.clamp_k_eq_gs128 [ OK ] test_kai_matmul_clamp_f32_qai8dxp4x8_qsi4c32p8x8_4x8x32_neon_i8mm.clamp_k_eq_gs128 (3 ms) [ RUN ] test_kai_matmul_clamp_f32_qai8dxp4x8_qsi4c32p8x8_4x8x32_neon_i8mm.m_clamp_k_eq_gs128 [ OK ] test_kai_matmul_clamp_f32_qai8dxp4x8_qsi4c32p8x8_4x8x32_neon_i8mm.m_clamp_k_eq_gs128 (5 ms) [----------] 6 tests from test_kai_matmul_clamp_f32_qai8dxp4x8_qsi4c32p8x8_4x8x32_neon_i8mm (121 ms total) ``` * [Exeprimental] Kleidi: rename arg name for packing functions * [Experimental] Change kernel cmake_out dir to avoid conflict
yanbing-j
pushed a commit
to yanbing-j/ao
that referenced
this pull request
Dec 9, 2024
* Revert Generate Behavior for non-Flamingo Models * Simplify
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Add kernel level tests, with basic cross compilation support.
Tested with S24 + r26c